Abstract: To enhance information security during transmission over public channels, images are frequently employed for binary data hiding. Nonetheless, data are vulnerable to distortion due to Joint Photographic Experts Group (JPEG) compression, leading to challenges in recovering the original binary data. Addressing this issue, this paper introduces a pioneering method for binary data hiding that leverages a combined spatial and channel attention Transformer, termed SCFformer, to withstand JPEG compression. This method employs a novel discrete cosine transform (DCT) quantization truncation mechanism during the hiding phase to bolster the stego image’s resistance to JPEG compression, using spatial and channel attention to conceal information in less perceptible areas, thereby enhancing the model’s resistance to steganalysis. In the extraction phase, the DCT quantization minimizes secret image loss during compression, facilitating easier information retrieval. The incorporation of scalable modules adds flexibility, allowing for variable-capacity data hiding. Experimental findings validate the high security, large capacity, and high flexibility of our scheme, alongside a marked improvement in binary data recovery post-JPEG compression, underscoring our method’s leading-edge performance.

SCFformer：一种基于空间通道融合Transformer的抗JPEG压缩的二进制数据隐藏方法

[1]Abood EW, Abdullah AM, Al Sibahe MA, et al., 2022. Audio steganography with enhanced LSB method for securing encrypted text with bit cycling. Bull Electr Eng Inform, 11(1):185-194.

[2]Baluja S, 2020. Hiding images within images. IEEE Trans Patt Anal Mach Intell, 42(7):1685-1697.

[3]Bhinder P, Singh K, Jindal N, 2018. Image-adaptive watermarking using maximum likelihood decoder for medical images. Multim Tools Appl, 77(8):10303-10328.

[4]Boehm B, 2014. StegExpose—a tool for detecting LSB steganography.

[5]Boroumand M, Chen M, Fridrich J, 2019. Deep residual network for steganalysis of digital images. IEEE Trans Inform Foren Secur, 14(5):1181-1193.

[6]Bui T, Agarwal S, Yu N, et al., 2023. RoSteALS: robust steganography using autoencoder latent space. Proc IEEE/CVF Conf on Computer Vision and Pattern Recognition Workshops, p.933-942.

[7]Chen BJ, Zhou CF, Jeon B, et al., 2018. Quaternion discrete fractional random transform for color image adaptive watermarking. Multim Tools Appl, 77(16):20809-20837.

[8]Guan ZY, Jing JP, Deng X, et al., 2023. DeepMIH: deep invertible network for multiple image hiding. IEEE Trans Patt Anal Mach Intell, 45(1):372-390.

[9]Guo LJ, Ni JQ, Shi YQ, 2012. An efficient JPEG steganographic scheme using uniform embedding. Proc IEEE Int Workshop on Information Forensics and Security, p.169-174.

[10]Guo MH, Xu TX, Liu JJ, et al., 2022. Attention mechanisms in computer vision: a survey. Comput Vis Med, 8(3):331-368.

[11]Harish NJ, Kumar BBS, Kusagur A, 2013. Hybrid robust watermarking techniques based on DWT, DCT and SVD. Int J Adv Electr Electron Eng, 2(5):137-143.

[12]Hayes J, Danezis G, 2017. Generating steganographic images via adversarial training. Proc 31^st Int Conf on Neural Information Processing Systems, p.1951-1960.

[13]Holub V, Fridrich J, Denemark T, 2014. Universal distortion function for steganography in an arbitrary domain. EURASIP J Inform Secur, 2014:1.

[14]Horé A, Ziou D, 2010. Image quality metrics: PSNR vs. SSIM. Proc 20^th Int Conf on Pattern Recognition, p.2366-2369.

[15]Huang XH, Deng ZF, Li DD, et al., 2023. MISSFormer: an effective Transformer for 2D medical image segmentation. IEEE Trans Med Imaging, 42(5):1484-1494.

[16]Isac B, Santhi V, 2011. A study on digital image and video watermarking schemes using neural networks. Int J Comput Appl, 12(9):1-6.

[17]Jassim FA, 2013. A novel steganography algorithm for hiding text in image using five modulus method. Int J Comput Appl, 72(17):39-44.

[18]Jing JP, Deng X, Xu M, et al., 2021. HiNet: deep image hiding by invertible network. Proc IEEE/CVF Int Conf on Computer Vision, p.4713-4722.

[19]Kim D, Shin C, Choi J, et al., 2023. Diffusion-Stego: training-free diffusion generative steganography via message projection.

[20]Kishore V, Chen XY, Wang Y, et al., 2022. Fixed neural network steganography: train the images, not the network. Proc 10^th Int Conf on Learning Representations, p.1.

[21]Li S, Zhang XP, 2019. Toward construction-based data hiding: from secrets to fingerprint images. IEEE Trans Image Process, 28(3):1482-1497.

[22]Li ZZ, Yang XY, Shen KQ, et al., 2023. Adversarial feature hybrid framework for steganography with shifted window local loss. Neur Netw, 164:358-369.

[23]Liu Z, Lin YT, Cao Y, et al., 2021. Swin Transformer: hierarchical vision Transformer using shifted windows. Proc IEEE/CVF Int Conf on Computer Vision, p.9992-10002.

[24]Lu W, Sun W, Lu HT, 2009. Robust watermarking based on DWT and nonnegative matrix factorization. Comput Electr Eng, 35(1):183-188.

[25]Lu W, Zhang JH, Zhao XF, et al., 2021. Secure robust JPEG steganography based on autoencoder with adaptive BCH encoding. IEEE Trans Circ Syst Video Technol, 31(7):2909-2922.

[26]Luo YJ, Zhou TQ, Liu F, et al., 2023. IRWArt: levering watermarking performance for protecting high-quality artwork images. Proc ACM Web Conf, p.2340-2348.

[27]Mallika, Ubhi JS, Aggarwal AK, et al., 2022. Neural style transfer for image within images and conditional GANs for destylization. J Vis Commun Image Represent, 85:103483.

[28]Mou C, Xu YM, Song JC, et al., 2023. Large-capacity and flexible video steganography via invertible neural network. Proc IEEE/CVF Conf on Computer Vision and Pattern Recognition, p.22606-22615.

[29]Shang F, Lan YH, Yang JH, et al., 2023. Robust data hiding for JPEG images with invertible neural network. Neur Netw, 163:219-232.

[30]Singh SP, Bhatnagar G, 2018. A new robust watermarking system in integer DCT domain. J Vis Commun Image Represent, 53:86-101.

[31]Singh SP, Rawat P, Agrawal S, 2012. A robust watermarking approach using DCT-DWT. Int J Emerg Technol Adv Eng, 2(8):300-305.

[32]Tao JY, Li S, Zhang XP, et al., 2019. Towards robust image steganography. IEEE Trans Circ Syst Video Technol, 29(2):594-600.

[33]Wei P, Li S, Zhang XP, et al., 2022. Generative steganography network. Proc 30^th ACM Int Conf on Multimedia, p.1621-1629.

[34]Xu GS, Wu HZ, Shi YQ, 2016. Structural design of convolutional neural networks for steganalysis. IEEE Signal Process Lett, 23(5):708-712.

[35]Ye J, Ni JQ, Yi Y, 2017. Deep learning hierarchical representations for image steganalysis. IEEE Trans Inform Foren Secur, 12(11):2545-2557.

[36]Ye YX, Shan J, Bruzzone L, et al., 2017. Robust registration of multimodal remote sensing images based on structural similarity. IEEE Trans Geosci Remote Sens, 55(5):2941-2958.

[37]Yu XZ, Chen KJ, Wang YF, et al., 2020. Robust adaptive steganography based on generalized dither modulation and expanded embedding domain. Signal Process, 168:107343.

[38]Zhang CN, Benz P, Karjauv A, et al., 2020. UDH: universal deep hiding for steganography, watermarking, and light field messaging. Proc 34^th Conf on Neural Information Processing Systems, p.10223-10234.

[39]Zhang KA, Cuesta-Infante A, Xu L, et al., 2019. SteganoGAN: high capacity image steganography with GANs.

[40]Zhang Y, Luo XY, Yang CF, et al., 2015. A JPEG-compression resistant adaptive steganography based on relative relationship between DCT coefficients. Proc 10^th Int Conf on Availability, Reliability and Security, p.461-466.

[41]Zhang Y, Luo XY, Wang JW, et al., 2021. Image robust adaptive steganography adapted to lossy channels in open social networks. Inform Sci, 564:306-326.

[42]Zhao ZZ, Guan QX, Zhang H, et al., 2019. Improving the robustness of adaptive steganographic algorithms based on transport channel matching. IEEE Trans Inform Foren Secur, 14(7):1843-1856.

[43]Zhu JR, Kaplan R, Johnson J, et al., 2018. HiDDeN: hiding data with deep networks. Proc 15^th European Conf on Computer Vision, p.682-697.